Breast tissue is very sensitive to ionizing radiation due to the presence of reproductive hormones, including estrogen. In the present pilot study, the efficiency of mammography X-rays to induce DNA double strand breaks (DSB) in mammary epithelial cells was investigated. For this, freshly resected healthy breast tissue was irradiated with 30 kV mammography X-rays in the dose range 0-500 mGy (2, 4, 10, 20, 40, 100 and 500 mGy). Breast specimens were also irradiated with identical doses of ⁶⁰Co γ-rays as a radiation quality standard. With the γH2AX-foci assay, the number of DNA DSB induced by radiation were quantified in the mammary epithelial cells present in breast tissue. Results indicated that foci induced by 30 kV X-rays and γ-rays followed a biphasic linear dose-response. For 30 kV X-rays, the slope in the low dose region (0-20 mGy) was 8.71 times steeper compared with the slope in the higher dose region (20-500 mGy). Furthermore, compared with γ-rays, 30 kV X-rays were also more effective in inducing γH2AX-foci. This resulted in a relative biological effectiveness (RBE) value of 1.82 in the low dose range. In the higher dose range, an RBE close to 1 was obtained. In conclusion, the results indicated the existence of a low dose hypersensitive response for DSB induction in the dose range representative for mammography screening, which is probably caused by the bystander effect. This could affect the radiation risk calculations for women participating in mammography screening.

This paper presents an analysis to develop a subjective state-of-knowledge probability distribution of a dose and dose-rate effectiveness factor for use in estimating risks of solid cancers from exposure to low linear energy transfer radiation (photons or electrons) whenever linear dose responses from acute and chronic exposure are assumed. A dose and dose-rate effectiveness factor represents an assumption that the risk of a solid cancer per Gy at low acute doses or low dose rates of low linear energy transfer radiation, RL, differs from the risk per Gy at higher acute doses, RH; RL is estimated as RH divided by a dose and dose-rate effectiveness factor, where RH is estimated from analyses of dose responses in Japanese atomic-bomb survivors. A probability distribution to represent uncertainty in a dose and dose-rate effectiveness factor for solid cancers was developed from analyses of epidemiologic data on risks of incidence or mortality from all solid cancers as a group or all cancers excluding leukemias, including (1) analyses of possible nonlinearities in dose responses in atomic-bomb survivors, which give estimates of a low-dose effectiveness factor, and (2) comparisons of risks in radiation workers or members of the public from chronic exposure to low linear energy transfer radiation at low dose rates with risks in atomic-bomb survivors, which give estimates of a dose-rate effectiveness factor. Probability distributions of uncertain low-dose effectiveness factors and dose-rate effectiveness factors for solid cancer incidence and mortality were combined using assumptions about the relative weight that should be assigned to each estimate to represent its relevance to estimation of a dose and dose-rate effectiveness factor. The probability distribution of a dose and dose-rate effectiveness factor for solid cancers developed in this study has a median (50th percentile) and 90% subjective confidence interval of 1.3 (0.47, 3.6). The harmonic mean is 1.1, which implies that the arithmetic mean of an uncertain estimate of the risk of a solid cancer per Gy at low acute doses or low dose rates of low linear energy transfer radiation is only about 10% less than the mean risk per Gy at higher acute doses. Data were also evaluated to define a low acute dose or low dose rate of low linear energy transfer radiation, i.e., a dose or dose rate below which a dose and dose-rate effectiveness factor should be applied in estimating risks of solid cancers.

The aim of this article is to offer a broader, mechanism-based, analytical tool than that used by (Rühm et al 2016 Ann. ICRP 45 262-79) for the interpretation of cancer induction relationships. The article explains the limitations of this broader analytical tool and the implications of its use in view of the publications by Leuraud et al 2015 (Lancet Haematol. 2 e276-81) and Richardson et al 2015 (Br. Med. J. 351 h5359). The publication by Rühm et al 2016 (Ann. ICRP 45 262-79), which is clearly work in progress, reviews the current status of the dose and dose-rate effectiveness factor (DDREF) as recommended by the ICRP. It also considers the issues which might influence a reassessment of both the value of the DDREF as well as its application in radiological protection. In this article, the problem is approached from a different perspective and starts by commenting on the limited scientific data used by Rühm et al 2016 (Ann. ICRP 45 262-79) to develop their analysis which ultimately leads them to use a linear-quadratic dose effect relationship to fit solid cancer mortality data from the Japanese life span study of atomic bomb survivors. The approach taken here includes more data on the induction of DNA double strand breaks and, using experimental data taken from the literature, directly relates the breaks to cell killing, chromosomal aberrations and somatic mutations. The relationships are expanded to describe the induction of cancer as arising from radiation induced cytological damage coupled to cell killing since the cancer mutated cell has to survive to express its malignant nature. Equations are derived for the induction of cancer after both acute and chronic exposure to sparsely ionising radiation. The equations are fitted to the induction of cancer in mice to illustrate a dose effect relationship over the total dose range. The 'DDREF' derived from the two equations varies with dose and the DDREF concept is called into question. Although the equation for acute exposure can be used to analyse atomic bomb survivor data, the fitting is dominated by the quadratic dose component. Thus, little useful information can be derived about the linear dose component which is important for the derivation of low dose rate risk. The ICRP are advised to derive the risk at low dose rates from epidemiological studies of, for example, worker populations, together with information from cellular radiation biological research.